DFI Magazine September/October 2020 - 84

* An arbitrary deformation limit such as
"2 inches of displacement" is not
meaningful unless it is specified exactly
which component of the system it
applies to - the roadway, top of
foundations, maximum ground displacement, etc.

S3 software discussion

Engineering that was made available
through collaboration with Silas Nichols,
P.E., of Federal Highway Administration
(FHWA) and Derrick Dasenbrock, P.E., of
FHWA (formerly with MNDOT).
Geomaterial properties were based on
those described in the Barr Engineering
report, with slight modifications to result in
a long-term drained FOS of approximately
1.0 considering a hydrostatic groundwater
table (in the real landslide, artesian
groundwater and rapid drawdown conditions were identified as causative mechanisms). The proposed stabilization scheme
investigated by the software providers
consisted of three rows of 8 ft (2.4 m)
diameter drilled shafts spaced 40 ft (12.1 m)
in-plane and 24 ft (7.3 m) out-of-plane,
with 16 ft (4.9 m) of embedment in the Red
Lake Falls Formation till that underlies the
soft clay layer where sliding occurs.
Presentations by Rocscience, Itasca,
Gregory Geotechnical and Bentley covered
a broad range of analysis methods, ranging
from 2D LEM SSA with separate or
integrated laterally-loaded pile analysis to
3D continuum numerical modeling
approaches based on digital topographic
surface models of the site. The presenters
provided additional insights into various
aspects of the system performance, such as
the effect of the shaft embedment length in
the till, the choice of appropriate p-y curves
(a nontrivial challenge when simulating
drained clay behavior), and schemes to
enhance the stabilizing action of the shafts,
such as connecting the rows with grade
beams to achieve frame action or adding
micropiles.

84 * DEEP FOUNDATIONS * SEPT/OCT 2020

Two findings were consistent from all
four presentations - that the shafts would
undergo a flexural failure without
mobilizing the full structural shear strength
of the structural section, and that the
amount of ground deformation required to
develop the flexural failure mechanism is
far above the amount of deformation that
would be acceptable for a slope supporting
a highway. From these findings it was clear
that modeling the shaft resistance as its
structural shear strength simply does not
provide a realistic understanding of the
actual slope behavior.

Lessons Learned
From the collective findings of the software
providers, several lessons and conclusions
were drawn during the closing panel discussion. The finding that the available
resistance was limited by slope deformation or a "serviceability" condition
rather than by reaching a structural or
geotechnical strength limit state highlighted several key points:
* You would never know this if you
simply assigned the structural shear
strength as a resisting force!
* FOS alone is not sufficient to quantify
whether the predicted slope performance is acceptable. Project owners or
engineers acting as the owner's representative are encouraged to specify performance criteria in the form of acceptable
deformation limits in addition to
factor(s) of safety. LRFD is appealing in
this regard because service and strength
limit states are inherently separate concepts; it also provides a rational framework for identifying factored demands
appropriate for LRFD structural design
of the foundations.

* Anticipated slope displacement, for
example, the amount of movement
associated with a just-stable slope at an
FOS of unity, is different from the
fictitious concept of the amount of
displacement required to mobilize a
strength limit state that corresponds to
an FOS greater than 1.0.
The detailed descriptions of the analysis
procedures also made it clear that, no
matter what approach is being used in
terms of 2D versus 3D, LEM versus
continuum numerical modeling, etc., the
problem cannot be understood without
explicitly analyzing the structural and
laterally loaded pile behavior. Hopefully
this will encourage practitioners to seek
out and implement knowledge on these
topics, or more effectively collaborate with
structural designers.
Finally, it was noted that combining a
deformation-based laterally loaded pile
analysis with LEM SSA violates the
fundamental premise of the LEM, which is
simply not capable of providing information about the amount of deformation
that accompanies a failure mechanism.
Practitioners need to keep in mind that
they are "breaking the rules" when using
this combined approach and to proceed
with caution. Slope stability FOS is not a
panacea, but rather a concept relating
demand to resistance, and should be
considered as one component of a robust
engineering solution to slope stabilization
projects.

Join Us at S3 2021!
The next S3: Slopes, Slides and Stabilization conference is being held August 3-5,
2021, at the Hotel Kabuki in San Francisco
next year. I'm excited!
DFI Magazine September/October 2020

Table of Contents for the Digital Edition of DFI Magazine September/October 2020
